Global Parameter Fitting Across Complementary Experiments

23 concepts17 papers1 proposal

When probing the fundamental laws of physics, researchers frequently encounter a frustrating barrier: parameter degeneracy. You build a multi-million dollar detector, collect pristine data, and find that your results can be explained equally well by two entirely different physical scenarios.

In isolation, a single experiment often maps out a "valley" of likely solutions in parameter space rather than a single point. To break these degeneracies, physicists use global parameter fitting across complementary experiments. By statistically fusing data from different detector materials, particle sources, or physical channels, they can intersect these independent solution valleys, collapsing broad uncertainties into precise measurements.

The Geometry of Degeneracy

To understand why global data fusion is necessary, imagine trying to determine two unknown physical constants, $x$ and $y$. Your experiment is highly sensitive to their sum, measuring $x + y = 10$ with extreme precision. On a graph of $x$ versus $y$, your allowed parameter space is a narrow diagonal band. No matter how many more events your detector collects, you cannot isolate the exact values of $x$ or $y$ because they are perfectly degenerate along that line.

To break the deadlock, you need an experiment with a different geometric orientation in parameter space—one that measures $x - y = 2$. When you combine the two datasets, the intersection of their orthogonal constraint bands pinpoints the exact solution ($x=6, y=4$).

This geometric intersection is the core mechanism behind multi-experimental fusion. As Rosner et al. demonstrated in charm physics and Ahmed et al. showed with solar neutrinos, integrating independent experimental yields into a single statistical model resolves deep parameter ambiguities that local datasets cannot handle alone.

Rotating the Constraint: Complementary Target Materials

In the search for Dark Matter, Weakly Interacting Massive Particles (WIMPs) are expected to scatter off atomic nuclei inside underground detectors. However, a light WIMP with a large scattering cross-section can produce a nuclear recoil spectrum nearly identical to that of a heavy WIMP with a small cross-section.

How do we break this mass-versus-cross-section degeneracy? We change the target nucleus.

Cross-Section
     ^
     |      \  Xenon Constraint
     |       \_______
     |       /       \
     |      /         \       Germanium Constraint
     |     /     *     \_____/
     |    /   True WIMP  _____/
     |   /     Value    /
     |  /______________/
     |
     +-----------------------------------> WIMP Mass

The kinematics of a WIMP-nucleus collision depend directly on the mass of the target atom. As explored by Pato et al. and Strege et al., combining data from detectors utilizing different elements—such as Xenon, Germanium, and Argon—rotates the degeneracy contours in the parameter space. What looks like an ambiguous smear to a Xenon detector becomes a sharp, well-defined mass peak when cross-referenced with Germanium data, yielding up to a twofold improvement in parameter reconstruction.

A similar principle applies to observing coherent elastic neutrino-nucleus scattering (CEvNS). By combining data from detectors with different nuclear masses, such as the Argon and Cesium-Iodide targets within the COHERENT experiment, Cadeddu et al. were able to extract significantly tighter constraints on fundamental neutrino properties.

Exploiting Physical Asymmetries: Neutrinos vs. Antineutrinos

In neutrino physics, a major goal is determining whether neutrinos violate Charge-Parity (CP) symmetry, which could explain why the universe is dominated by matter rather than antimatter. However, the mathematical probability of a neutrino changing flavor depends simultaneously on the CP-violating phase ($\delta_{CP}$) and the "mass hierarchy" (whether the third neutrino mass state is the heaviest or lightest).

A single beam of neutrinos cannot easily disentangle these two effects. To solve this, experiments like NOvA run two distinct operational modes: a neutrino beam and an antineutrino beam.

As explained by Acero et al., CP violation affects neutrinos and antineutrinos in exactly opposite ways. If a physical parameter enhances neutrino appearance, it suppresses antineutrino appearance. By performing a joint analysis of both beams, the authors could decouple the mass hierarchy from the CP phase, transforming a highly correlated parameter space into a robust, unambiguous measurement.

Combining Beams and Nature: Accelerator vs. Atmospheric Data

Another powerful axis of complementarity lies in the choice of particle source. Long-baseline accelerator experiments shoot a controlled, narrow-energy beam of neutrinos through the Earth to a distant detector. This setup is ideal for pinning down specific oscillation minima with high precision.

However, these artificial beams have a fixed baseline. In contrast, naturally occurring atmospheric neutrinos are produced by cosmic rays hitting the upper atmosphere. They rain down on detectors from all directions, providing a vast spread of path lengths (from 20 km for neutrinos produced overhead to 12,000 km for those traveling through the Earth's core) and a wide range of energies.

Accelerator Data:   Fixed Baseline  ---> High Precision at Specific Energy
Atmospheric Data:   Varying Paths   ---> High Sensitivity to Earth Matter Effects
Joint Fit:          Combines Both   ---> Breaks Mass Hierarchy Degeneracy

As demonstrated by Adamson et al. and Adamson et al. using the MINOS experiment, combining accelerator and atmospheric datasets creates a powerful synergy. The accelerator data anchors the basic oscillation parameters, while the atmospheric data—having traversed varying amounts of the Earth's mantle—introduces matter-induced MSW effects that definitively break the degeneracy surrounding the neutrino mass hierarchy.

The Statistical Engine: Joint Likelihoods

True data fusion is not merely the overlapping of final published intervals. If two experiments independently calculate their parameters and you simply multiply their final probabilities, you risk introducing severe biases. This is because different experiments often share underlying systematic uncertainties, such as theoretical models of neutrino flux, nuclear cross-sections, or dark matter halo densities.

To correctly fuse data, physicists construct a joint global likelihood function. This statistical engine fits the primary physics parameters alongside dozens of "nuisance parameters" representing the systematic uncertainties of every included experiment.

$$\mathcal{L}_{\text{global}}(\theta; \eta) = \mathcal{L}_{\text{Exp A}}(\theta; \eta_{\text{shared}}, \eta_A) \times \mathcal{L}_{\text{Exp B}}(\theta; \eta_{\text{shared}}, \eta_B)$$

Where $\theta$ represents the fundamental physics parameters of interest, and $\eta$ represents the systematic nuisance parameters.

This joint framework allows for cross-calibration. For example, in searching for hypothetical sterile neutrinos, Adamson et al. and Adamson et al. combined reactor antineutrino disappearance (Daya Bay, Bugey-3) with accelerator muon neutrino disappearance (MINOS, MINOS+). Because the reactor experiments tightly constrain the initial electron-flavor flux, the accelerator experiments can look for anomalies without being bottlenecked by their own flux uncertainties.

We see this ultimate expression of data fusion in landmark joint analyses across entirely separate experimental collaborations, such as the combined oscillation analysis from the T2K and NOvA experiments (Abubakar et al.), the blending of reactor and spallation-source data (Atzori Corona et al.), and the unification of collider data with indirect astronomical observations (Bertone et al.). By treating the world's detectors not as isolated projects, but as components of a single distributed instrument, global parameter fitting uncovers the precise architecture of the subatomic world.

History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_multi_detector_data_fusion_7acdc155,
  title = {multi-detector data fusion},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/multi-detector-data-fusion/rdr_7acdc155448d4e4d9e81767323a5cb0f},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}

This exact version (version 1, frozen text):

@misc{lacuna_multi_detector_data_fusion_7acdc155_v1,
  title = {multi-detector data fusion},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/multi-detector-data-fusion/rdr_7acdc155448d4e4d9e81767323a5cb0f/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake